J . Am. Chem. Soc., Vol. 103, No. 10, 1981 2909
Additions and Corrections
amine19 produced the silyl ether of cholesterol. Deprotection of the 3P-hydroxyl group yielded cholesterol which was spectroscopically identical with an authentic sample (I3C NMR, 360MHz 'H N M R , and IR). The alkylideneoxirane 5 provided the complementary proof for our methodology. Addition of lithium methylcyanocuprate to 5 resulted in the formation of equal amounts of 1,4 and 1,2 adducts, 720and 8,21respectively. The unanticipated 1,2 adduct (8) can only be rationalized this time by the differences in the steric bulk of the alkylidene substituents (Le., methyl vs. isohexyl). Comparison of the 360-MHz 'H N M R spectra of the enantiomeric adducts 6 and 7 revealed a clear distinction in the absorptions for the respective (2-21 methyl groups (0.97 for 6 and 1.1 1 for 7). Intermediate 7 was then converted to the known isocholesterolZ2 by the same sequence of steps used for the preparation of cholesterol. As a final check on the purity of the C-20 epimers produced in this study, we subjected the dimethyl ethyl silyl (DMES) ethers of compounds 6, 7, 9, 10, cholesterol, and isocholesterol to G C analysis.23 While the retention times for the DMES ethers of 6 and 7 were very close, the ethers of compounds 9, 10, cholesterol, and isocholesterol separated well enough that it was possible to detect less than 1% epimeric contaminants. The G C analyses24 confirmed that our synthetic products, 6 , 7 , 9 , 10, cholesterol, and isocholesterol were >99.5% pure epimers at C-20. In conclusion, the 1,Ctrans addition of alkyl cyanocuprates to alkylideneoxiranes of sterols provides the only stereospecific methodology for the concomitant introduction of the C-21 asymmetric center and the 15P-hydroxyl group. This synthetic (19) Ireland, R. E.; Muchmore, D. C.; Hengartner, U. J . Am. Chem. SOC. 1972, 94, 5098. (20) 7: [a]270 -87.6' (c 0.21, CHCI,); IH N M R (360 MHz) 6 1.06 (s, 3 H, H-18) 1.11 (s, 3 H, H-19), H-21 overlapped with 1.11, 3.43-3.52 (m, 1 H , H-31, 4.44-4.48 (br, 1 H, H-15), 5.30-5.38 (br, 1 H, H-6), 5.53 (d, 1 H, J = 2.20 Hz).
strategy should be applicable to a wide variety of functionalized sterol side chains and functionalized D rings of steroids.
Acknowledgment. We are pleased to acknowledge support of this research from the National Cancer Institute (NIH) under Grant CA 22237. The National Science Foundation is acknowledged for providing funds to purchase a Bruker 360-MHz N M R spectrometer. We also thank the G. D. Searle Co. for a generous gift of dehydroepiandrosterone. Supplementary Material Available: Experimental details and characterization data for compounds 2-10 are available upon request (1 1 pages). Ordering information is given on any current masthead.
(21) 8: [a]270 -66.9' (c 0.22, CHCI,); 'H N M R (360 MHz) 6 1.03 (s, 3 H, H-lS), 1.05 (d, 3 H, J = 7.32 Hz, H-l6CH,), 1.18 ( s , 3 H, H-19), 3.42-3.52 (m, 1 H , H-3), 3.80 (d, 1 H, J = 4.39 Hz, H-20), 5.00-5.08 (br, 1 H, H-20), 5.28-5.36 (br, 1 H, H-6). (22) Nes, W. E.; Varkey, T. E.; Krevitz, K. J . Am. Chem. SOC.1977, 99, 260. Koreeda, M.; Koizumi, N . Tetrahedron &it. 1978, 1641. See also 5a. Isocholesterol: mp 151-153 'C; 'H N M R (360 MHz) 6 0.66 (s, 3 H, H-18), 0.80 (d, 3 H, J = 6.59 Hz, H-21), 0.99 ( s , 3 H, H-19); 13CNMR (90 MHz) 140.83, 121.70.71.86, 56.90, 55.91, 50.32,42.43, 39.81, 39.53, 37.37, 36.62, 35.82, 35.23, 32.05, 31.80, 28.10, 28.00, 24.27, 24.03, 22.75, 22.66, 21.21, 19.43. 18.72. 12.18. (23) For G C analysis of 20R and 2 0 s steroidal acetates, see: Schow, S. R.; McMorris, T. C. J . Org. Chem. 1979, 44, 3760. (24) The DMES ethers of 6, 7, 9, 10, cholesterol, and isocholesterol were prepared at room temperature with neat (dimethylethylsilyl)imidazole (Miyazaki, H.; Ishibashi, M.; Itoh, M.; Nambura, T. Biomed. Mass Spectrom. 1977, 4 , 23). G C analyses were carried out on a Shimadzu TP-MI gas chromatograph (FID) with a 1.8-m (5-mm 0.d.) column of 3% SE-30. Column temperature: 320 'C for ethers of 6, 7, 9, 10; 300 "C for ethers of cholesterol and isocholesterol. Gas flow: nitrogen 40 mL/min; hydrogen 40 mL/min. Retention times relative to T H F solvent: DMES ether of 6, 9.6 min; 7, 9.4 min; 9, 8.2 min; 10, 11.2 min; cholesterol, 10.6 min; isocholesterol, 9.8 min.
Additions and Corrections Studies on the Reaction Mechanism of the Photocyclization of N-Aryl Enamines. Dependence of Quantum Yields on Back and Side Reactions. [ J . Am. Chem. SOC.1980,102,6098]. THOMAS WOLFF*and REINHARDT WAFFENSCHMIDT, Gesamthochschule Siegen, Physikalische Chemie, D-5900 Siegen 21, West Germany. Page 6099, column 1, first paragraph: The N M R data for 1-(N-methylani1ino)-1-phenyl-1-propene(3) should read: 'H N M R (CDClJ 1.7 (3 H, doublet), 3.1 (3 H, singlet), 6.1 (1 H, quartet), 6.6-7.4 ppm (10 H, multiplet). The authors are indebted to Professor H. Ahlbrecht, Giessen, for detecting and communicating the error. Synthesis of P-Lactams from Substituted Hydroxamic Acids [ J . Am. Chem. SOC.1980, 102, 70261. M. J. MILLER,*P. G. MATTINGLY, M. A. MORRISON,and J. F. KERWIN,JR., Department of Chemistry, University of Notre Dame, Notre Dame, Indiana 46556. The products from the intermolecular N-alkylation of 0-benzyl hydroxamates (RCONHOCH,Ph) with alcohols ( R 2 0 H ) in the presence to DEAD/PPh, (last four entries in Table I) have been subsequently shown to be the 0-alkyl isomers
as reported. The products from the intermolecular alkylation of 0-acylhydroxamates and the intramolecular alkylations to give P-lactams are correctly assigned.
I5N Nuclear Magnetic Resonance Spectroscopy. Products and Rearrangementsin the Reaction of p-ToluenesulfonylAzide-3-I5N with the Sodium Salt of p-Toluenesulfonamide. An in Situ I5N NMR Study [ J . Am. Chem. SOC.1980, 102, 23641. CARLA Contribution No. 61 12 from the CASEWITand J. D. ROBERTS,* Gates and Crellin Laboratories of Chemistry, California Institute of Technology, Pasadena, California 91 125 Page 2365, column 2: Figure l a should appear as shown below.
1V WNOl
0-R2 R-C=N-OCH,
Ph
and not the N-alkyl isomers O R I1
R-C-N-
a
I
OCH, Ph
YnhWI"
